Pathology Associates Of Lexington, P.A.

PROSTATE BIOPSIES: OPTIMIZED CANCER

DETECTION AND STAGING

ERVIN B. SHAW, M. D., BEVERLY W. DANIEL, M. D., ELIZABETH D. WOFFORD, M. D., JOHN B. CARTER, M. D. [agar embedding process explained & photos]

The importance of early diagnosis of prostate cancer has recently received much attention both in the medical literature^1,2 and lay media.³ While some calculate that the cost to screen and then manage every diagnosable prostate cancer would be too burdensome, our impression is that a great majority of individual patients desire early diagnosis; and more treatment options are available at early diagnosis. In view of such possible conflicting motivations, we determined to do a systematic study of the effectiveness of a maximal-detection prostate biopsy processing system devised by one of us (EBS) in late 1991.⁴

The frequency and comprehensiveness of needle biopsies of the prostate gland has increased significantly since 1990 following the introduction of PSA screening and the availability of a reliable, automated biopsy gun technique.³ While the surgical pathology evaluation of these needle biopsies is definitive in the diagnosis of prostatic cancer, there is significant variation in surgical pathology laboratories as to the methods and thoroughness with which these biopsies are studied. The majority of prostate biopsies are done in doctors' offices, and biopsy specimens are often processed by remote commercial laboratories. In many urology practices, the multiple biopsies are submitted in a single container (or possibly divided into right-side and left-side containers) and processed as a cluster of tissue fragments, most often with two to three levels of microscopic sections studied. We introduced a system to separately identify each needle biopsy specimen whether dealing with lesser or greater numbers of specimens, but geared toward the sextant technique of prostate gland biopsies. This sextant approach (three spaced biopsies ... apex, mid, base ... per each side of the gland, six biopsies in all for each case) is used in cases of an elevated PSA without a definite palpable lesion or definite target lesion by ultrasound.⁵

Effectiveness of the surgical pathology process was addressed in the form of two questions: (1) How much surgical pathology work was necessary to detect all cancers contained in the biopsy specimens?⁴ (2) Did the sextant pattern plus this thorough process produce the highest diagnostic yield of cancers?

SURGICAL PATHOLOGY THOROUGHNESS

Traditionally and by common practice, surgical pathology processing of needle biopsies has involved sectioning of a single core biopsy at two levels of section on two slides. With the advent of multiple (often sextant) needle biopsies of the prostate gland, surgical pathology processing remained relatively unchanged. Multiple biopsies not separately identified with lateralizing labels are often processed altogether in a single cluster of needle biopsy cores. Increasingly, right and left-sided biopsies are segregated into separate bottles, each of the multiple biopsies processed in a non-oriented cluster in a single block cut at two levels of section on two slides. An informal verbal survey of surgical pathology departments in early 1993 indicated that most departments reviewed multiple frames of section on two or three slides. The bundle or wad of needle biopsy cores precluded the ability to view an entire plane of section of each needle biopsy. Rarely was there any attempt to spatially orient the sections.

In an effort to systematize and optimize the information to be gained from prostatic needle biopsies, we developed a technique for specimen orientation of transrectal ultrasound-guided (TRUS) needle biopsies using disposable biopsy guns.⁴ Labeling and orientation of each biopsy is maintained from the biopsy needle to the pathology slide by use of directional orientation blot papers (Fig. 1) and agar pre-embedding of formalin fixed biopsy cores. Blot papers are made from hand-towel paper, the best being paper which is quick-absorbing yet firm enough to maintain its shape when wet. We prefer Kimberly Clark's Surpass hand towels. Even a fragmented biopsy maintains perfect orientation on a labeled blot paper. Keeping the blot paper label in mind, the blot paper is mildly firmly touched by the operator or an assistant onto the biopsy gun needle containing the tissue core, the core instantly absorbing to the paper (Fig. 2), the paper then folded into the labeled specimen container holding the 10 percent neutral buffered formalin fixative (we recommend against any other type of fixative). With careful attention to maintaining this orientation, the cores are removed onto the metal surgical pathology dissecting table and arranged in relation to pre-cut, colored-agar orientation markers. Biopsies and markers are then pre-embedded in clear, melted agar which rapidly hardens (Fig. 3).⁶ At least three biopsy cores can be agar pre-embedded in the same plane of sectioning in one paraffin block.⁴ Orientation is maintained within the paraffin block (and thereafter on the histological slides [Fig. 4]) by the presence of the colored-agar markers, enabling precise determination as to the localization (that is, capsular end vs. deep end of the tissue core) and measured extent of any neoplasia identified. Location and inferred estimate of tumor extent or volume may facilitate planning and clinico-pathologic correlation of such procedures as endorectal coil MRI staging studies or therapeutic options. Interval paraffin ribbons can be saved for ancillary studies, including immuno-microscopy and DNA profiles [error: we later found that DNA profiles require much thicker sections].

We performed a look-back review of our entire experience of 540 consecutive unselected patient cases processed during a 12-month period (1993) to determine the optimum number of levels of section per block necessary for detection of all neoplasia actually contained in the sum of the 18 gauge biopsy cores.⁴ Prior to and during the study, an average of 10 slides representing evenly spaced step-cut sections through the full thickness of the needle biopsy core was made for each case with a range of seven to 16 slides per block, two or three frames of tissue on each microscopic slide. This original processing effort entailed exhaustive review of the 540 patient cases; hence, we assumed that all carcinomas actually contained in biopsy cores had been detected... that none were missed, there being no significant amount of tissue remaining in the paraffin blocks. Using from one to eight biopsies per case, 105 (19 percent) cases were carcinomatous. 29 of the malignant cases (28 percent) had only a single, minute (3mm or less) focus of neoplasia in only one of the biopsy cores.

Our look-back analysis of this exhaustive processing technique determined that all of the observed cancers would have been detect ed by the study of five slides (15 frames of section) of level-plane, spatially oriented needle biopsy sections. So, how much work is necessary on a sextant series? To provide a margin of safety, we determined as policy to prepare six step-sectioned slides (two to three frames of section per slide) from each paraffin block, step-sectioning the needle biopsy cores completely through (six slides, each at a separate plane of section) and saving intervening ribbons until the case interpretation was completed. Paraffin ribbons from unusual cases could be saved indefinitely.

A clinical records review of the 29 cases with a single minute focus of carcinoma revealed that 11 underwent radical prostatectomy. Complete surgical pathology examination of all 11 surgically resected prostate glands confirmed the diagnosis of adenocarcinoma, all tumors being greater than 9mm in diameter or multi-focal in their involvement of the prostate gland. Very importantly, three of the eleven cases showed tumor extending to the surgical resection margins. Thus, in agreement with others,⁷ the detection of a single minute tumor focus using multiple (sextant) needle biopsies cannot be assumed to reflect insignificant carcinoma as such may well represent a small sampling of a much larger tumor that may be clinically significant. So-called "minute carcinoma" in a sextant biopsy series has a very significant chance of not being minute carcinoma in the resected prostate gland!

THOROUGHNESS PLUS SEXTANT PATTERN

Others⁸ have indicated a higher detection rate of cancer when using a sextant or systematic patterned biopsy approach to cases with PSA elevation but without evidence of a target nodule. A review of 355 consecutive, unselected patients from three area urology office general practices biopsied with 18 gauge needles during 1995 indicated a significantly increased yield (39 to 43 percent) if from three to seven biopsies were taken, as opposed to two or less (24 percent) (Table 1). Our review could not distinguish cases as being with or without clinical target lesions. 12.8 percent of the 196 patients undergoing sextant biopsies contained cancer in only one of the six biopsies (Table 1). While one might intuitively expect more biopsies to have a greater diagnostic yield, we saw about a 40 percent yield so long as one utilized at least three biopsies, a result probably attributable to the heterogeneous mixture of biopsy indications in our unselected series. If the utility of screening and early biopsy diagnosis of prostate cancer is accepted as a mark of medical effectiveness, our data suggests that there is a significant potential of a false-negative biopsy study if less than a three-biopsy series is performed.

In summary it is important to emphasize that minute carcinoma in a needle biopsy may not be minute clinically. It only requires a measured zone of cancer 4 mm in length in one biopsy core to define "clinically significant" (volume > 0.5cc) cancer.⁷ Hence, optimally sensitive processing of prostate needle biopsies done according to the method of highest yield is essential to maximized detection. Experience related to the 540 patient series indicates that six slides carrying six spaced levels through the full thickness of the spatially oriented, single-plane biopsy cores will detect all diagnosable neoplasia within these biopsies. A less thorough examination hazards the possibility of missing an important malignancy. A sextant series requires only two paraffin blocks.

The spatial orientation of the needle biopsies offers the ability to roughly locate the tumor site as to the base, mid, and apical portions of the right or left prostate lobes and to very roughly estimate tumor size (at least as to a determination of clinically significant size).⁷ The biopsy often contains the capsular boundary, and our process allows one to portray the relationship of a cancerous segment to the capsule. The capsular end of a biopsy core may sometimes be discerned grossly by the presence of a thin, elastic tail of periprostatic areolar tissue stringing from that end of the biopsy core. The "capsule" of the gland is often visible microscopically as a slightly increased peripheral loosening of fibromuscular tissue, usually in contact with some microscopic adipose tissue. Capsular involvement by tumor on a needle biopsy most likely is, thereby, a true positive sign reflecting the real status of tumor within the prostate gland. The presence of ganglion cells is another indicator of "capsular" area. Of course, absence of such as capsular or perineural tumor involvement on needle biopsy cannot be assumed to be a true rule-out result as the amount of sampling is so small.

Though not the focus of this paper, it has not escaped our notice that pre-analytical (prior to specimen receipt in the lab) factors are crucial to maximized cancer detection and interpretation. These include: quality PSA determinations, the urologist's choice of the number of biopsies, biopsying of the most cancer-prevalent plane (in the absence of a target lesion, para-peripheral rather than paramedial),⁷ and snug positioning of the biopsy gun needle tip at the glandular-periglandular interface (not pre-penetrated into the gland) prior to triggering the biopsy. Such positioning accomplishes two important objectives. Firstly, the capsular region is, thereby, likely to be contained in the biopsy sample and be recognizable histologically. Secondly, each biopsy thereby consists predominately of a maximized length of actual prostate gland tissue, a longer cumulative length of multiple biopsy cores being intuitively expected to detect more cancer than shorter cumulative lengths. Concern for cancer location misinterpretation due to the relatively tangential direction of the biopsy plane could be allayed were a modification to be invented in the biopsy gun for an angled biopsy path more directly ... less tangentially ... into the gland. Finally, this fully oriented biopsy approach allows both a focused approach to the surgical pathology processing of any subsequent radical prostatectomy specimen, especially in cases of pre-operative neo-adjuvant total androgen blockade where documentation of the residual cancer can sometimes be so problematic.⁹ And a radiologico-pathological correlation of findings is possible for a more optimized endorectal coil MRI interpretation as we are providing in our hospital. Though the reader may infer that our process is overly compulsive and too time consuming, once set into place, it is actually a smoothly repetitious, relatively easy approach to a significantly demanding area of surgical pathology in view of the advantages gained.

REFERENCES:

1. Optenberg SA, Thompson IM. Economics of screening for carcinoma of the prostate. Urol Clin North Am. 17:719-737, 1990.

2. Slawin KM; Ohori M, Dillioglugil 0, Scardino, PT. Screening for prostate cancer: An analysis of the early experience. CA-A Ca J Clin. 45(3):134-147, 1995.

3. Kolata G. Advances in detection create dilemma on prostate cancer. The New York Times; Al, 1993.

4. Shaw EB, Wofford ED, Carter, J.: Processing prostate needle biopsy specimens for 100% detection of Carcinoma. Am J Clin Pathol. 103(4):507,1995.

5. Hodge KK, McNeal JE, Terris MK, Stamey TA. Random systematic versus directed ultrasound guided transrectal core biopsies of the prostate. J. Urol. 142:71-74,1989.

6. Watson CG, Johnson JM, Shaw EB. Correct orientation of specimens for histologic processing - Preliminary embedding in agar, Am J Derm Path. 5:165-167, 1983.

7. Stamey TA. Making the most out of six systematic sextant biopsies. Urol. 45:2-12, 1995.

8. Terris MK, McNeal JE, Stamey TA. Detection of clinically significant prostate cancer by transrectal ultrasound-guided systematic biopsies. J. Urol. 148:829-832, 1992.

9. Armas OA, Aprikian AG, Reuter VE, "et al." Clinical and pathobiological effects of neoadjuvant total androgen ablation therapy on clinically localized prostatic adenocarcinoma. Am J Surg Pathol. 18:979-991, 1994.